Medical Physiology, 3rd Edition

Motor Units and Muscle Function

Chapter 9 deals with the cellular and molecular physiology of skeletal muscle contraction. In this subchapter, we examine the integration of these cellular and molecular elements into the contraction of a whole muscle.

The motor unit is the functional element of muscle contraction

The motor unit (see pp. 228–229) is the functional unit of skeletal muscle and consists of a single motor neuron and all the muscle fibers that it activates. A typical skeletal muscle such as the biceps brachii receives innervation from ~750 somatic motor neurons.

When the motor neuron generates an action potential, all fibers in the motor unit fire simultaneously. Thus, the fineness of control for movement varies with the innervation ratio (see pp. 228–229)—the number of muscle fibers per motor neuron. imageN60-1 The small motor units that are recruited during sustained activity contain a high proportion of type I muscle fibers (see p. 249), which are highly oxidative and resistant to fatigue. The type IIa motor units (see p. 249) have larger innervation ratios, contract faster, and have less oxidative capacity and fatigue resistance than type I units. Still larger motor units—recruited for brief periods for rapid, powerful activity—typically consist of type IIx muscle fibers; these have less oxidative capacity (i.e., are more glycolytic) and are much more susceptible to fatigue than type IIa muscle. In practice, it may be best to view muscle fiber types and motor units as a continuum rather than distinct entities.


Innervation Ratio

Contributed by Steven Segal, Emile Boulpaep, Walter Boron

motor unit is a motor neuron and all of the muscle fibers that it innervates (see pp. 228–229). The innervation ratio quantifies the motor unit. The innervation ratio is the number of muscle fibers innervated by a single motor neuron. Thus, large motor units have high innervation ratios.

Muscles controlling the very fine movements of the eye (e.g., rectus lateralis) have innervation ratios of only a few muscle fibers. At the other extreme, motor units in the thigh (e.g., quadriceps muscle) and calf (e.g., gastrocnemius muscle) often contain thousands of muscle fibers and are involved in such powerful activities as sprinting and jumping. In the muscles involved in hand movements, motor units have innervation ratios that vary from <100 fibers (e.g., interosseous muscles controlling fine movements of the fingers) to >1000 fibers (e.g., forearm muscles controlling such coarse movements as grasping).

Within a whole muscle, muscle fibers of each motor unit intermingle with those of other motor units so extensively that a volume of muscle containing 100 muscle fibers is innervated by terminals from perhaps 50 different motor neurons. However, each muscle fiber is innervated by only one motor neuron. Within some muscles, the fibers of a motor unit are constrained to discrete compartments. This anatomical organization enables different regions of a muscle to exert force in somewhat different directions, which enables more precise control of movement.

Muscle force rises with the recruitment of motor units and an increase in their firing frequency

During contraction, the force exerted by a muscle depends on (1) how many motor units are recruited, and (2) how frequently each of the active motor neurons fires action potentials. Motor units are recruited in a progressive order, from the smallest (i.e., fewest number of muscle fibers) and therefore the weakest motor units to the largest and strongest. This intrinsic behavior of motor-unit recruitment is known as the size principle and reflects inherent differences in the biophysical properties of respective motor neurons. For a given amount of excitatory input (i.e., depolarizing synaptic current; Isyn in Fig. 60-1), a neuronal cell body with smaller volume and surface area has a higher membrane input resistance. Therefore, the depolarizing voltage rises to threshold more quickly in a neuron with a smaller neuronal cell body than in a neuron with a larger cell body (see Fig. 60-1). Because the neurons with the small cell bodies tend to innervate a small number of slow-twitch (type I) muscle fibers, the motor units with the greatest resistance to fatigue are the first to be recruited. Conversely, the neurons with the larger cell bodies tend to innervate a larger number of fast-twitch (type II) fibers, so the largest and most fatigable motor units (type IIx) are the last to be recruited (e.g., during peak levels of force production). Because the relative timing of action potentials in different motor units is asynchronous, the force developed by individual motor units integrates into a smooth contraction. As a muscle relaxes, the firing of respective motor units diminishes in reverse order.


FIGURE 60-1 Size principle for motor units. Small motor neurons are more excitable, conduct action potentials more slowly, and excite fewer fibers that tend to be slow twitch (type I). Conversely, large motor neurons are less excitable, conduct action potentials more rapidly, and excite many fibers that tend to be fast twitch (type II). EPSP, excitatory postsynaptic potential; Rm, membrane resistance; Vm, membrane potential. (Adapted from Kandel ER, Schwartz JH, Jessell TM: Principles of Neural Science, 4th ed. New York, McGraw-Hill, 2000.)

At levels of force production lower than the upper limit of recruitment, gradations in force are accomplished via concurrent changes in the number of active motor units and the firing rate of those that have been recruited—rate coding. Once all the motor units in a muscle have been recruited, any further increase in force results from an increase in firing rate. The relative contribution of motor-unit recruitment and rate coding varies among muscles. In some cases, recruitment is maximal by the time muscle force reaches ~50% of maximum, whereas in others, recruitment continues until the muscle reaches nearly 90% of maximal force.

Not only do the intrinsic membrane properties of motor neurons (i.e., the size principle) affect motor neuron firing, but other neurons that originate in the brainstem project to the motor neurons and release the neuromodulatory neurotransmitters serotonin and norepinephrine (see Fig. 13-7A and B). For example, this neuromodulatory input, acting on the motor neurons of small, slow-twitch motor units, can promote self-sustained levels of firing of the motor neurons during the maintenance of posture. In contrast, the withdrawal of this excitatory neuromodulatory input during sleep promotes muscle relaxation. Thus, the brainstem can control the overall gain of a pool of motor neurons.

Compared with type I motor units, type II units are faster and stronger but more fatigable

Within a given motor unit, each muscle fiber is of the same muscle fiber type. As summarized in Tables 9-1 and 9-2, the three human muscle fiber types—type I, type IIa, and type IIx—differ in contractile and regulatory proteins, the content of myoglobin (and thus color) and of mitochondria and glycogen, and the metabolic pathways used to generate ATP (i.e., oxidative versus glycolytic metabolism). Physical training can modify these biochemical properties, which determine a range of functional parameters, including (1) speeds of contraction and relaxation, (2) maximal force, and (3) susceptibility to fatigue (Fig. 60-2).


FIGURE 60-2 Properties of fiber types. The top row shows the tension developed during single twitches for each of the muscle types; the arrows indicate the time of the electrical stimulus. The middle row shows the tension developed during an unfused tetanus. The bottom row shows the degree to which each fiber type can sustain force during continuous stimulation. Note that the time scales become progressively larger from the top to bottom rows, with a break in the bottom row. Also, the tension scales become progressively larger from left (fewer fibers per motor unit) to right (more fibers per motor unit). These results are from experiments performed on cat gastrocnemius muscle, which has type IIb fibers (instead of type IIx fibers as in humans). (Data from Burke RE, Levine DN, Tsairis P, et al: J Physiol 234:723–748, 1977.)

In response to an action potential evoked through the motor axon, slow-twitch (type I) motor units (see Fig. 60-2A, top) require relatively long times to develop moderate levels of force and return to rest. In contrast, fast-twitch (types IIa and IIx) motor units exhibit relatively short contraction and relaxation times and develop higher levels of force (see Fig. 60-2B and C, top). Accordingly, during repetitive stimulation (middle row of Fig. 60-2), slow-twitch motor units summate to a fused tetanus at lower stimulation frequencies than do fast-twitch motor units. Indeed, the α motor neurons in the spinal cord that drive slow motor units fire at frequencies of 10 to 50 Hz, whereas those that drive fast motor units fire at 30 Hz to >100 Hz.

The maximal force that can develop per cross-sectional area of muscle tissue is constant across fiber types (~25 N/cm2). Therefore, the ability of different motor units to develop active force is directly proportional to the number and diameter of fibers each motor unit contains. In accord with the innervation ratios of motor units, peak force production (middle row of Fig. 60-2) increases from type I motor units (used for fine control of movement) to type II motor units (recruited during more intense activities).

The susceptibility to fatigue of a motor unit depends on the metabolic profile of its muscle fibers. The red type I muscle fibers have a greater mitochondrial density and can rely largely on the aerobic metabolism of carbohydrate and lipid as fuel because they are well supplied with capillaries for delivery of O2 and nutrients. Type I motor units, although smaller in size (and innervation ratio), are recruited during sustained activity of moderate intensity and are highly resistant to fatigue (see Fig. 60-2A, bottom). In contrast, the larger type II motor units are recruited less often—during brief periods of intense activity—and rely to a greater extent on short-term energy stores (e.g., glycogen stored within the muscle fiber). Among type II motor units, type IIa motor units have intermediate innervation ratios, a greater mitochondrial density, a larger capacity for aerobic energy metabolism, a greater O2 supply, and a higher endurance capacity, and hence are classified as fast fatigue-resistant (bottom of Fig. 60-2B). In contrast, type IIx motor units have the highest innervation ratios and a greater capacity for rapid energy production through nonoxidative (i.e., anaerobic) glycolysis, and thus can produce rapid and powerful contractions. However, type IIx units tire more rapidly and are therefore classified as fast fatigable (see Fig. 60-2C, bottom).

As external forces stretch muscle, series elastic elements contribute a larger fraction of total tension

As sarcomeres contract, some of their force acts laterally—through membrane-associated and transmembrane proteins—on the extracellular matrix and connective tissue that surrounds each muscle fiber. Ultimately, the force is transmitted to bone, typically (but not always) through a tendinous insertion. The structural elements that transmit force from the cross-bridges to the skeleton comprise the series elastic elements of the muscle and behave as a spring with a characteristic stiffness. Stretching resting muscle causes passive tension to increase exponentially with length (see Fig. 9-9C). Thus, muscle stiffness increases with length. During an isometric contraction (see pp. 237–238), when the external length of a muscle (or muscle fiber) is held constant, the sarcomeres shorten at the expense of stretching the series elastic elements. An isometric contraction can occur at modest levels of force development, such as holding a cup of coffee, as well as during maximal force development, such as when opposing wrestlers push and pull against each other, with neither gaining ground. Physical activity typically involves contractions in which muscles are shortening and lengthening, as well as periods during which muscle fibers are contracting isometrically.

During cyclic activity such as running, muscles undergo a stretch-shorten cycle that may increase total tension while decreasing active tension. For example, as the calf muscles relax as the foot lands and decelerates, the series elastic elements of the calf muscles (e.g., the Achilles tendon, connective tissue within muscles) are stretched and develop increased passive tension (see Fig. 9-9C). Thus, when the calf muscles contract to begin the next cycle, they start from a higher passive tension and therefore use a smaller increment in active tension to reach a higher total tension. imageN60-2 This increased force helps to propel the runner forward.


Effect of Stretch on the Active Tension of Skeletal Muscle

Contributed by Emile Boulpaep, Walter Boron

As shown in Figure 9-9D, the active tension of skeletal muscle is maximal at a sarcomere length with an optimal overlap of the thin and thick filaments. At relatively low initial lengths of the muscle (e.g., at 70 on the x-axis of Fig. 9-9C, D), the active tension is relatively low. Prestretching the muscle to a greater initial length produces a greater active tension … up to a point (i.e., 100 on the x-axis). Further increases in passive length actually produce a decrease in active tension (e.g., 130 on the x-axis). Of course, the total tension increases continuously from the lowest to the highest muscle lengths.

In a concentric contraction (e.g., climbing stairs), the force developed by the cross-bridges exceeds the external load, and the sarcomeres shorten. During a concentric contraction, a muscle performs positive work (force × distance) and produces power (work/time; see p. 240). As shown in Figure 9-9E, the muscle achieves peak power at relatively moderate loads (30% to 40% of isometric tension) and velocities (30% to 40% of maximum shortening velocity). The capacity of a muscle to perform positive work determines physical performance. For example, a stronger muscle can shorten more rapidly against a given load, and a muscle that is metabolically adapted to a particular activity can sustain performance for a longer period of time before succumbing to fatigue.

In an eccentric contraction (e.g., descending stairs), the force developed by the cross-bridges is less than the imposed load, and the sarcomeres lengthen. During an eccentric contraction, the muscle performs negative work, thereby providing a brake to decelerate the applied force being applied, and absorbs power. Eccentric contractions can occur with light loads, such as lowering a cup of coffee to the table, as well as with much heavier loads, such as decelerating after jumping off a bench onto the floor. At the same absolute level of total force production, eccentric contractions—with increasingly stretched sarcomeres—develop less active tension than do concentric or isometric contractions. Conversely, the passive tension developed by the series elastic elements makes a greater contribution to total tension. As a result, the maximum tension generated eccentrically can be greater than that generated isometrically. When the external force stretches the muscle sufficiently, all the tension is passive and the limit is the breaking point (see Fig. 19-9B) of the series elastic elements. Thus, eccentric contractions are much more likely than isometric or isotonic contractions to damage muscle fibers and connective tissue, as occurs with the common injury of a ruptured Achilles tendon.

The action of a muscle depends on the axis of its fibers and its origin and insertion on the skeleton

In addition to the contractile and metabolic properties of muscle fibers discussed above, two anatomical features determine the characteristics of the force produced by a muscle.

The first anatomical determinant of muscle function is the arrangement of fibers with respect to the axis of force production (i.e., the angle of pennation). With other determinants of performance (e.g., fiber type and muscle mass) being equal, muscles that have a relatively small number of long fibers oriented parallel to the axis of shortening (e.g., the sartorius muscle of the thigh; Fig. 60-3A, top) shorten more rapidly. Indeed, the more sarcomeres are arranged in series, the more rapidly the two ends of the muscle will approach each other. In contrast, muscles that have many short fibers at an angle to the axis (e.g., the gastrocnemius muscle of the calf; Fig. 60-3A, bottom) develop more force. Indeed, the greater the number of fibers (and sarcomeres) in parallel with each other, the greater the total cross-sectional area for developing force.


FIGURE 60-3 Determinants of the mechanical action of a muscle.

The second anatomical determinant of limb movement consists of the locations of the origin and insertion of the muscle to the skeleton. Consider, for example, the action of the brachialis muscle on the elbow joint. The distance between the insertion of the muscle on the ulna and the joint's center of rotation is D, which may be 5 cm. The torque that the muscle produces on the joint is the component of total muscle force that is perpendicular to the ulna, multiplied by D (see Fig. 60-3B). An equivalent definition is that torque is the product of the total muscle force multiplied by the moment arm, which is the length of the line segment that runs perpendicular to the muscle and through the center of rotation (see Fig. 60-3B). As we flex the elbow, the moment arm is constantly changing and muscle force changes as well. For this joint, we achieve maximum torque at 60 degrees of flexion.

Fluid and energetically efficient movements require learning

To perform a desired movement—whether playing the piano or serving a tennis ball—the nervous system must activate a combination of muscles with the appropriate contractile properties, recruit motor units in defined patterns, and thereby create suitable mechanical interactions among body segments. When we perform movements with uncertainty—as in learning a new skill—actions tend to be stiff because of concurrent recruitment of motor units in antagonistic muscles that produce force in opposite directions. Such superfluous muscle fiber activity also increases the energy requirements for the activity. Even in someone who is skilled, the fatigue of small motor units leads to the recruitment of larger motor units in the attempt to maintain activity, but with loss of fine control and greater energy expenditure. With learning, recruitment patterns become refined and coordinated, and muscle fibers adapt to the task. Thus, movements become fluid and more energetically efficient, as exemplified by the movements of highly trained musicians and athletes, who can make difficult maneuvers appear almost effortless.

Strength versus endurance training differentially alters the properties of motor units imageN60-3


Endurance (“Aerobic”) versus Strength (“Anaerobic”) Training

Contributed by Emile Boulpaep, Walter Boron

The distinction between endurance training (e.g., long-distance running) and strength training (e.g., weightlifting) refers to the manner of exercise. Endurance training involves performing a lower-intensity activity for a longer period. Strength training, in contrast, involves performing a high-intensity activity for shorter periods.

The distinction between aerobic and anaerobic refers to the metabolic pathway that the muscles primarily use to regenerate ATP. Thus, in aerobic exercise, the cells regenerate ATP primarily by using oxidative phosphorylation. In anaerobic exercise—of course, no one exercises in the absence of oxygen!—the cells regenerate ATP primarily by using anaerobic glycolysis, generating lactic acid in the bargain.

The firing pattern of the α motor neuron—over time—ultimately determines the contractile and metabolic properties of the muscle fibers in the corresponding motor unit. This principle was demonstrated by cross-innervation experiments in which investigators cut the motor nerve to a muscle composed primarily of fast motor units and switched it with the motor nerve of a muscle composed primarily of slow motor units. imageN60-4 As the axons regenerate and the muscles recover contractile function over several weeks, the fast muscle becomes progressively slower and more fatigue resistant whereas the slow muscle becomes faster and more susceptible to fatigue. Varying the pattern of efferent nerve impulses via long-term stimulation of implanted electrodes elicits similar changes in muscle properties. A corollary of this principle is that physical activity leads to adaptation only in those motor units that are actually recruited during the activity.


Cross-Innervation Experiments

Contributed by Steven Segal

The principle described in the text—that it is the type of motor neuron that determines the properties of the muscle fibers the neuron innervates—was demonstrated in experiments reported by Buller, Eccles, and Eccles in a classic paper published in 1960.


Buller AJ, Eccles JC, Eccles RM. Interactions between motoneurones and muscles in respect of the characteristic speeds of their responses. J Physiol. 1960;150:417–439.

The effects of physical activity on motor-unit physiology depend on the intensity and duration of the exercise. In general, sustained periods of activity of low to moderate intensity performed several times per week—endurance (aerobic) training—result in a greater oxidative capacity of muscle fibers and are manifested by increases in O2 delivery, capillary supply, and mitochondrial content (see pp. 1219–1222). These adaptations reduce the susceptibility of the affected muscle fibers to fatigue (see pp. 1212–1213). The lean and slender build of long-distance (i.e., endurance) runners reflects the abundance of highly oxidative type I and IIa muscle fibers of relatively small diameter, promoting O2 and CO2 diffusion between capillaries and mitochondria for high levels of aerobic energy production. Further, the high ratio of surface area to volume of the slender body also facilitates cooling of the body during prolonged activity and in hot environments.

In contrast, brief sets of high-intensity contractions performed several times per week—strength (anaerobic) training—result in type IIx motor units that can produce more force and can shorten against a given load at greater velocity by increasing the amount of contractile protein. The hypertrophied muscles of sprinters and weightlifters exemplify this type of adaptation, which relies more on rapid, anaerobic sources of energy production (see p. 1209).